Publication bias accounts for SSRI benefit in autism
HealthDay News -- The selective publication of positive trial results may explain the reported benefits of selective serotonin reuptake inhibitors (SSRIs) for treating repetitive behaviors in autism spectrum disorders (ASD), researchers suggest.
SSRIs and the tricyclic antidepressant clomipramine (Anafranil) were associated with a modest but significant improvement in repetitive behaviors (P<0.005), results of a meta-analysis published online in Pediatrics indicate.
However, the improvements lost statistical significance after accounting for publication bias, Melisa Carrasco, PhD, of the University of Michigan in Ann Arbor, and colleagues reported.
"Without timely, transparent, and complete disclosure of trial results, it remains difficult to determine the efficacy of available medications," the researchers wrote.
Several previously published randomized clinical trials have explored SSRIs as a treatment option for children with ASDs, but findings about their usefulness have been unclear.
To explore the topic further, Carrasco and colleagues identified 10 completed randomized, double-blind, placebo-controlled trials: five published and five unpublished. All of the published studies and only one of the unpublished studies were included in the meta-analysis. Data for all trials were requested, but not supplied. A total of 365 patients were included in the final analysis of four SSRIs -- clomipramine, fluvoxamine (Luvox), fluoxetine (Prozac) and citalopram (Celexa).
A small but significant effect of SRI for the treatment of repetitive behaviors in ASD (standardized mean difference, 0.22; z score = 2.87; P<0.005) was observed. Significant evidence of publication bias was noted in all analyses. When the effect of publication bias was adjusted for, the benefit of SSRIs was no longer significant (standardized mean difference=0.12). There was no significant effect between type of medication, patient age, method of analysis, trial design or trial duration on reported SSRI efficacy.
In an accompanying editorial, Scott Denne, MD, of Indiana University School of Medicine in Indianapolis criticized what he called a "lack of timely and complete disclosure of trial findings," noting that it inhibits clinicians from making "rational informed decisions about the efficacy and risks" of SSRIs for treating children with ASDs.
He emphasized the importance of registering all clinical trials and the necessity of publishing or making any results, whether positive or negative, available to researchers and the public. Not doing so ultimately leads to unnecessary repetition, and the information learned not being used to guide therapy.
Furthermore, not disclosing all trial results falls short of meeting an obligation to the children and families who participate in clinical trials to ensure that their contributions are used to help all children.
"Only timely, complete, and readily available clinical trial results can meet this obligation. We have a viable mechanism (ClinicalTrials.gov) to post trial results, but have fallen far short of the goal," Denne wrote. "Addressing this deficiency will require a renewed commitment by clinical investigators, the NIH, the pharmaceutical industry and the FDA. It is time for urgent action."